Pose

[john-rocky]

@ambitious-octopus

I've added pose functionality to the app and committed to the pose branch.
This branch is based on the branch that
added human and segmentation.
As always, please let me know if there are any issues that need to be resolved during the integration process (e.g. maintaining model references).

Thanks in advance.

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

This pull request introduces numerous enhancements to the 'ultralytics/yolo-ios-app' project, improving model selection, updating dependencies, and adding various utility functions for better processing and handling of video frames and model outputs.

📊 Key Changes

• README Formatting: Improved README formatting for better readability.
• Project Configuration: Multiple updates in the Xcode project file including adding new source files and dependencies.
• Model Handling: Separate models for different tasks (detection, segmentation, human, pose).
• BoundingBoxView Enhancements: Added layers to manage visualization of bounding boxes and labels.
• Additional Utility Files: Introduced Colors.swift, HumanModel.swift, PostProcessPose.swift, PostProcessSegment.swift, PostProcessing.swift, TrackingModel.swift.
• UI Enhancements: Updated main storyboard for an improved UI layout, added landscape mode adjustments.
• Video Capture: Improved configuration and handling for video capture sessions.
• Task Management: Enhanced task segmentation through new task selection control and tracking switch.
• Performance Metrics: Improved frame per second (FPS) and inference time calculation and display.
• File Operations: Utility functions for saving and handling image files and text files.

🎯 Purpose & Impact

• Purpose:

  • Enhance Usability: Improved UI and added new functionalities for better app usability.
  • Broaden Model Coverage: Enable different YOLO models and segmentation tasks.
  • Improve Performance Tracking: Better FPS and inference time tracking for performance analysis.
  • Facilitate Debugging and Development: Enhanced logs, metrics, and file operations for developers.

• Impact:

  • Users: Provides a more user-friendly interface and supports more use-cases with multiple model selections.
  • Developers: Simplifies debugging and development with structured code and new utilities.
  • Performance: Optimized frame capturing and processing, offering smoother app experience.
  • Flexibility: Allows better real-time modifications and tracking during video capture.

Overall, these changes provide significant improvements in functionality, performance tracking, and usability of the YOLO iOS app, broadening its application potential and making it easier to use and develop. 🚀

[ambitious-octopus]

Hey @john-rocky , could you please review this PR and remove any unnecessary elements? I’d like to ensure we’re focusing solely on items related to pose estimation. Thanks!

[john-rocky]

@ambitious-octopus
Understood. I will modify this branch to a simple code that adds pose function to the main branch. Please wait a moment.

[john-rocky]

@ambitious-octopus
I removed unnecessary elements. Currently, this branch only adds the pause function to the main branch.

[ambitious-octopus]

Hey @john-rocky it seems that there are still some merge conflicts. Can you please resolve them?

[john-rocky]

@ambitious-octopus
I fixed the conflicts.
I tested this branch. This works fine.